Computer Power and Human Reason
Joseph Weizenbaum 1923-2008 Immigrated to American from Nazi Germany during at 13; became a professor emeritas of computer science at MIT (began there in 1964). Coded that famed DOCTOR: ELIZA program, the first convincing chatterbot. ELIZA was a language analysis program, produced in 2 tiers--a language analyzer and a script (a set of rules that defines the conversational context). She "like Eliza of Pygmalion fame it could be taught to speak increasingly well." (369). ELIZA was popular at MIT because "most other programs could not vividly demonstrate the information processing power of a computer" to visitors lacking computation or math skills and because it required interaction--perhaps not unlike [[Spacewar!|''Spacewar! ]]1962 functioned (370). The discussions regarding the "humanness" of this program and the eventual production of fully realized A.I., occupied much of Weizenbaum's later written work. He was often positioned as a couterpoint to Kenneth Colby, the Stanford psychiatrist who wanted to use computers in the theraputic situation, helped Weizenbaum with initial designs for ELIZA, and was the force behind the far more advanced chatterbot, PARRY. Weizenbaum's writing carries in intense humanism, and throughout his life he was concerned with the dire consequences of conflating human and machine. Weizenbaum's work was part of larger Cold War investments in computer science funded by the U.S. government, which were especially prevalent at MIT. ''Computer Power and Human Reason 1976 *this is not an entry on the full book, but on the cut from The New Media Reader* In this influential book, Weizenbaum uses the responses to ELIZA to argue that "a line dividing human and machine intelligence must be drawn," or else there is no point to understanding our humanity (371). Weizenbaum writes to address "my own shock ... by some people who insisted on misinterpreting a piece of work I had done" (368). ELIZA becomes a "parsimonious way" of identifying certain concerning issues (369). He defines his 3 shocks in relation to ELIZA: 1. Psychiatrists wanted to use it as an automated therapist. Weizenbaum believes this process requires empathy, an "engaged human being acting as a healer" (370) 2. People became emotionally involved with ELIZA and anthropomorphized her (the famous secretary story). Weizenbaum notes human attachment has trascended the machine (cars, computers) and can now be applied to the program; attachment doesn't require unique material presence. 3. People believed it showed a solution to the problem of computers understanding natural language. Weizenbaum asserts that it proves the importance of context. These 3 shocks in turn provoke a series of questions (371): *What are scientists responsibilities in regards to publicizing his work (by publishing the ELIZA script, Weizenbaum unwittingly opened the A.I. can of worms)? *To who is the scientist responsible? *What about the computer brings the view of man as machine to a new level of plausibility (that machines can now supposedly mimic our intellectual and not just physical function) Man as Clockwork Weizenbaum's philosophical extrapolation was that if man could not reasonably divide himself from machine, then man threatened to only see himself as mere clockwork, an organic mechanism that can be broken down and fully copied. While man does integrate machines into perceptual and kinesthetic habits, it is sloppy to presume a machine is analogous to how man thinks. This anxiety over viewing man as regulated and systematic brings up larger themes of Modernist functionality and the processes that ordered the Holocaust, which would reasonably occupy Weizenbaum's perspective. Rationality-as-Logicality Computers have reinforced and amplified the antecedent pressures the drive man's rationalistic view of the world and mechanistic view of himself. Thus, Weizenbaum's interest is not in ELIZA or even the computer, but in the larger questions and concerns that the computer amplifies. He articulates 2 camps in this discussion: computers do all, versus there must be a limit to computers. This debate is about whether or not "every aspect of human thought is reducible to a logical formalism ... entirely computable" (373). Prior to these technologies, the systems of thought that governed man were juridical, defining relations between men and nature. Religious systems are invested in man's autonomy and responsibility. Spiritual cosmologies of science, however, are "infected by the germ of logical necessity ... claim to say how things are and must necessarily be. They convert truth to provability" (not unlike the imperative of the Foucauldian truth-sky) (373). Regardless of how intelligent a computer may mange to be, some acts of thoughts ought to be the domain of a human. Weizenbaum identifies science as a addictive drug (and MIT as a "temple of technology") that sold us the age of rationality, but we have confused it with logicality. Weizenbaum writes: "the rationality-as-logicality equation, which the very success of science has drugged us into adopting as virtually an axiom, has led us to deny the very existence of human conflict, hence the very possibility of the collision of genuinely incommensurable human interests and of disparate human values, hence the existence of human values themselves" (the implication is that for a "logical" system, all disagreement is a communication problem--it suggests the Holocaust was not a matter of fundamentally incompatible human values) (374). This suggests logicality is an ideologicalsystem. Weizenbaum invests in human values, and says in no way can science disprove human values because if human values are illusory, so is sicnece, as it is knowledge garnered from man's own formal systems. "Belief in the rationality-logicality equation has corroded the prophetic power of language itself. We can could, but we are rapidly forgetting how to say what is worth counting and why" (375).